recognition of color morphs as separate species in ... - Springer Link

2 downloads 37 Views 1MB Size Report
Faria and Melo (2007) were the first to document the variation of general integumental color among individuals from the same species. According to them ...
Original article

Apidologie (2014) 45:641–652 * INRA, DIB and Springer-Verlag France, 2014 DOI: 10.1007/s13592-014-0280-7

Deceiving colors: recognition of color morphs as separate species in orchid bees is not supported by molecular evidence Bruno R. FERRARI , Gabriel A. R. MELO Laboratório de Biologia Comparada de Hymenoptera, Departamento de Zoologia, Universidade Federal do Paraná, Caixa Postal 19020, Curitiba, PR 81531-980, Brazil Received 24 September 2013 – Revised 3 January 2014 – Accepted 17 February 2014

Abstract – The orchid bees includes many species whose integument color is predominantly metallic, in particular those from the genus Euglossa. In the Atlantic Forest, some species exhibit color variation that seems to be linked with their distribution, with warmer colors (green, with coppery and reddish tones) in lower latitudes and cooler colors (blue, with purple shades) in higher latitudes. Contrasting taxonomic interpretations of this color variation lead to the recognition of distinct species for the different color morphs. In all three case studies investigated in this work (Euglossa iopoecila, Euglossa stellfeldi, and Euglossa townsendi), the different forms were sampled throughout their distribution. Thirty samples from 16 different localities were used, which had their DNA extracted, amplified, and sequenced for fragments of the genes cytochrome b (cytb) and cytochrome C oxidase subunit 1 (CO1). A concatenated data matrix, totaling 1,089 bp, was submitted for Bayesian and maximum likelihood analyses. The resulting phylogenetic trees, together with the estimated genetic distances between samples, led to the conclusion that the different color morphs in all the three case studies do not represent separate species, but only integumental color variations of their respective species. Atlantic Forest / Euglossini / mtDNA / molecular taxonomy

1. INTRODUCTION Orchid bees are a charismatic insect group in the Neotropical fauna, being widely known for their close association with flowers, most notably from Orchidaceae, that offer floral volatiles as reward. Many orchid bee species are also recognized for their brilliant metallic integumental coloration. In species of Euglossa, greenish coloration predominates, with variable tones and nuances ranging from copper (as in Euglossa pleosticta Dressler) to bluish (as in Euglossa viridis Perty). Many species, however, have their dominant integument color deep blue, Corresponding author: G. Melo, [email protected] Manuscript editor: Marina Meixner

even sometimes verging on purple tones, as in Euglossa iopyrrha Dressler (Dressler 1982a, b). Characters of integumental coloration have been used frequently by taxonomists within the genus Euglossa. These generally involve color of specific structures, such as the clypeus (Rebêlo and Moure 1996) and terminal terga (Dressler 1982a). Faria and Melo (2007) were the first to document the variation of general integumental color among individuals from the same species. According to them, Euglossa iopoecila Dressler, 1982, and Euglossa stellfeldi Moure, 1947, include both blue and green specimens throughout their distribution. In the same study, they also discuss Euglossa anodorhynchi Nemésio, 2006, a species proposed for its predominantly blue integument color, but which is very similar structurally to Euglossa townsendi Cockerell.

642

B.R. Ferrari and G.A.R. Melo

Faria and Melo (2007) also remarked that this kind of variation follows a latitudinal pattern, with green individuals occurring in the northern portions of their distribution and blue individuals occurring farther south along the Atlantic Forest, in eastern Brazil. Furthermore, specimens of E. iopoecila and E. stellfeldi with intermediate coloration (greenish-blue color) were documented at the middle of the species distribution (northeastern coast of São Paulo and southeastern coast of Rio de Janeiro), possibly revealing the existence of hybridization zones. Nemésio (2009), in a review of orchid bees of the Atlantic Forest, stated that coloration is genetically defined and should be given the same weight as other morphological characters already used in the group. He proposed that integument color should be recognized as a valid taxonomic character, strong enough to sustain the proposal of new species. Furthermore, he discussed the noticeable influence of humidity on coloration, which can be seen while drying and humidifying pinned specimens in collections. While drying, individuals tend to become more reddish and, after humidifying, they shift to more bluish tones. Although humidity’s influence was recognized, Nemésio rejected the notion that latitudinal humidity variation could be responsible for color variation within species. Based in these arguments, Nemésio (2009) proposed new species, some morphologically similar to those species cited by Faria and Melo (2007), basing taxonomic decisions mainly on color characters. The details involving the three case studies here discussed are summarized below. E. iopoecila case study Nemésio (2009) did not agree with the interpretation of Faria and Melo (2007) for E. iopoecila and argued that the northern green morph should be treated as a separate species, proposing for it the name Euglossa roubiki Nemésio, 2009. Under his view, the name E. iopoecila would apply only to the southern blue morph. E. stellfeldi case study Nemésio (2009) misapplied the name E. stellfeldi, using it for the species known as Euglossa annectans Dressler, 1982. For E. stellfeldi proper (sensu Faria &

Melo), he used two different names, Euglossa solangeae Nemésio, 2007, for the greenish blue morph found at the southern portion of the distribution, and reinstated the name Euglossa carinilabris Dressler, 1982, for the northern green morph. This latter name had been synonymized under E. stellfeldi by Faria and Melo (2007). E. townsendi case study Rebêlo and Moure (1996) interpreted E. townsendi, a species described originally from Mexico, as having a very wide distribution, ranging from Mexico to Argentina. Nemésio (2009), upon comparing the holotype of E. townsendi with similar females from southeastern Brazil, reached the conclusion that they could not be conspecific and proposed the name Euglossa aratingae for the latter population. Additionally, another recently described species, E. anodorhynchi Nemésio, 2006, ranging from coastal areas of Santa Catarina to southern São Paulo, presents an identical morphology to E. townsendi, except for its strong blue integument color, leading Faria and Melo (2007) to suggest that E. anodorhynchi could “represent only a color variation within a more widely distributed taxon.”

Considering the opposing taxonomic views, we conducted a study in which sequence data from two mitochondrial gene fragments were used to explore the boundaries between the color morphs of these three Euglossa cases in order to elucidate the identity of the biological units involved.

2. MATERIALS AND METHODS For extraction of molecular data, 41 recently collected males (maximum age of 4 years after collection) and one female from 18 localities in Brazil and Mexico were used. Different color specimens were used for each case, including blue, green, and, when available, intermediate color individuals (Figures 1, 2, and 3). The specimens were attracted using the protocols of Dodson et al. (1969), where artificial

Deceiving colors: recognition of color morphs as separate species in orchid bees

643

Figure 1. Map of eastern Brazil showing the geographic origin of the specimens of Euglossa iopoecila used in the present study. Original distribution of the Atlantic forest is indicated in darker gray. Different color morphs are illustrated at the right. scents are used to attract euglossine males that are captured with entomological nets. Specimen data

for the samples that were successfully amplified and sequenced are presented in Table I.

Figure 2. Map of eastern Brazil showing the geographic origin of the specimens of Euglossa stellfeldi used in the present study. Original distribution of the Atlantic forest is indicated in darker gray. Different color morphs are illustrated at the right.

644

B.R. Ferrari and G.A.R. Melo

Figure 3. Maps showing the geographic origin of the specimens of Euglossa townsendi used in the present study. Original distribution of the Atlantic forest in eastern Brazil is indicated in darker gray. Different color morphs are illustrated at the right.

Maps showing the geographic origin of the sequenced specimens (Figures 1, 2, and 3) and the known distribution of E. townsendi (including the taxa described as E. anodorhynchi and E. aratingae), based on specimens deposited in the entomological collection of the Universidade Federal do Paraná (Figure 4), were prepared in Diva-Gis 7.5.0. The genetic material was extracted from mesosomal muscular tissue (with four exceptions, BFt02, BFt03, BFt07, and BFt08, where genetic material was extracted from leg maceration) with the EZ-DNA Extraction Kit (Biosystems®, Brazil). For polymerase chain reactions (PCRs), we used Mastercycler Personal and Mastercycler Gradient thermocyclers (Eppendorf) and the following reagents: 2–5 μL of DNA sample, 2.5 μL of 10× PCR buffer (Invitrogen), 2.5 μL of MgCl2 (25 mM), 0.4 μL of dNTPs (25 mM), 1 μL of each primer (25 mM), 0.2 μL of Platinum® Taq DNA Polymerase (Invitrogen), and with the final volume of 25 μL obtained by addition of Milli-Q autoclaved water. Two gene fragments were amplified: cytochrome c oxidase subunit 1 (CO1) and cytochrome b (cytb) with the usage of the universal primers HCO and LCO (CO1; Folmer et al. 1994) and mtD-26 and

mtD-28 (cytb; Simon et al. 1994). These fragment positions were referenced with the complete mitochondrion genome of Apis mellifera Linnaeus (Crozier and Crozier 1993), to identify CO1 (1837-2494, 657 bp) and cytb (11428-11860, 432 bp). The successful amplified samples were sent to Macrogen in South Korea for purification and sequencing with an ABI 3730xl DNA Analyzer (Applied Biosystems) sequencer. Sequence data from additional species of Euglossina and of other subtribes in Apini were included in the analyses in order to provide topological structuring for the resulting phylograms and to allow for simultaneous analysis of the three cases here investigated. These sequences where taken from GenBank, and their data are detailed in Table II. All the sequences were edited using Geneious R6 (Biomatters Ltd., 2005–2013) and aligned in the same software by using the MAFFT multiple aligner v 1.3 plugin (Katoh et al. 2002) to produce a concatenated matrix with both genes. From this matrix, genetic pairwise p distances were calculated using MEGA v5.2 (Tamura et al. 2011). The Bayesian phylogenetic reconstruction was performed through CIPRES Science Gateway online

Green

Green

Blue

Intermediate

BFi05

BFi06

BFi07

Blue

BFa02

BFi04

Blue

BFa01

Green

Green

BFt11

BFi03

Green

BFt09

Green

Green

BFt08

BFi02

Green

BFt07

Green

Green

BFt03

BFi01

Green

BFt02

E. iopoecila

Green

BFt01

E. townsendi

Color morph

Code

Case study

accession numbers

Brazil, Bahia Ilhéus, UESC, 14.798° S, 39.172° W, 15 January 2010, G. Melo Brazil, Bahia Ilhéus, UESC, 14.798° S, 39.172° W, 15 January 2010, G. Melo Brazil, Rio de Janeiro, Com. Levy Gasparian, 21.995° S, 43.325° W, 28 October 2010, C.M. Maia & L.M. Santos Mexico, Veracruz, Los Tuxtlas, 18.5° N, 95.016° W, June 2009, D. Schorkopf Mexico, Campeche, Atasta, 18.55° S, 92.166° W, June 2009, T. Eltz Brazil, Minas Gerais, Uberlândia, Faz. Água Limpa, 19.166° S, 48.4° W, 19 July 2008, T.M.S. Mesquita Brazil, Minas Gerais, Bandeira, Res. Biodiversitas, 15.783° S, 40.516° W, 19 January 2011, G. Melo Brazil, Paraná, Paranaguá, F.E. Palmito, 25.566° S, 48.516° W, 9 December 2009, L.P. Amaral Neto Brazil, Paraná, Antonina, RPPN Rio Cachoeira, 25.378° S, 48.696° W, 1 February 2007, A.J.C. Aguiar Brazil, Bahia, Santa Teresinha, Serra Jibóia, 12.853° S, 39.475° W, 8 June 2007, J.A. Rafael & F.F. Xavier Brazil, Espírito Santo, Sooretama, Res. Vale, 19.152° S, 40.019° W, 6 January 2011, B.R. Ferrari & E. Ferrari Brazil, Bahia, Camacã, Serra Bonita, 15.393° S, 39.564° W, 21 January 2011, G. Melo Brazil, Espírito Santo, Sooretama, Rebio Sooretama, 18.992° S, 40.129° W, 7 January 2011, B.F. Ferrari & E. Ferrari Brazil, Rio de Janeiro, Rio das Ostras, Rebio União, 22.427° S, 42.043° W, 12 January 2011, B.R. Ferrari & E. Ferrari Brazil, Rio de Janeiro, Paraty, PN Serra da Bocaina, 23.208° S, 44.822° W, 17 January 2011, B.R. Ferrari & E. Ferrari Brazil, São Paulo, Ubatuba, Estrada Rio Escuro, 23.475° S, 45,140° W, 19 January 2011, B.R. Ferrari & E. Ferrari

Specimen data

KF443101

KF443105

KF443100

KF443103

KF443099

KF443127

KF443131

KF443126

KF443129

KF443125

KF443124

KF443107

– KF443098

KF443109

KF443108

KF443110

KF443086

KF443085

KF443087

KF443114



KF443113



KF443112

KF443115



KF443088



KF443111

– KF443089

cytb

CO1

GenBank accession no.

Table I. List of specimens of Euglossa included in this study, separated by taxonomic case, with their respective voucher code, color morph, collecting data, and GenBank

Deceiving colors: recognition of color morphs as separate species in orchid bees

645

E. stellfeldi

Case study

Table I (continued)

Intermediate

Green

Blue

Green

Green

BFs05av

BFs05v

BFs06a

BFs06v

BFs07

Blue

BFi14

Blue

Blue

BFi13

BFs05a

Blue

BFi12

Blue

Green

BFi10

BFs04

Green

BFi09

Green

Green

BFi08

BFs01

Color morph

Code

Brazil, São Paulo, Ubatuba, Estrada Rio Escuro, 23.475° S, 45,140° W, 19 January 2011, B.R. Ferrari & E. Ferrari Brazil, Rio de Janeiro, Guapimirim, PN Serra dos Órgãos, 22.493° S, 42.998° W, 13 January 2011, B.R. Ferrari & E. Ferrari Brazil, Bahia, Camacã, Faz. Paris, 15.425° S, 39.497° W, 22 January 2011, G. Melo Brazil, São Paulo, Sete Barras, Faz. Morro do Capim, 22.366° S, 47.966° W, 13 January 2009, Mattozo & Thomaz Brazil, Paraná, Antonina, RPPN Rio Cachoeira, 25.378° S, 48.696° W, 23 June 2007, Faria Jr, Weiss & Mattozo Brazil, Paraná, Antonina, RPPN Rio Cachoeira, 25.378° S, 48.696° W, 24 November 2007, Faria Jr, Weiss & Mattozo Brazil, Bahia, Camacã, Faz. Santa Bárbara, 15.383° S, 39.533° W, 14 January 2007, G. Melo & Carvalho Brazil, Paraná, Antonina, RPPN Rio Cachoeira, 25.378° S, 48.696° W, 27 October 2007, Faria Jr, Weiss & Mattozo Brazil, São Paulo, Ubatuba, Estrada Rio Escuro, 23.475° S, 45,140° W, 19 January 2011, B.R. Ferrari & E. Ferrari Brazil, São Paulo, Ubatuba, Estrada Rio Escuro, 23.475° S, 45,140° W, 19 January 2011, B.R. Ferrari & E. Ferrari Brazil, São Paulo, Ubatuba, Estrada Rio Escuro, 23.475° S, 45,140° W, 19 January 2011, B.R. Ferrari & E. Ferrari Brazil, Rio de Janeiro, Paraty, APA Cairuçú, 23.269° S, 44.667° W, 18 January 2011, B.R. Ferrari & E. Ferrari Brazil, Rio de Janeiro, Paraty, APA Cairuçú, 23.269° S, 44.667° W, 18 January 2011, B.R. Ferrari & E. Ferrari Brazil, Bahia, Camacã, Serra Bonita, 15.393° S, 39.564° W, 21 January 2011, G. Melo

Specimen data

KF443134 KF443133

– –

KF443121

KF443120

KF443094 KF443095

KF443122

KF443119

KF443118

KF443117

KF443123

KF443096

KF443093

KF443092

KF443091

KF443097

KF443116

KF443135



KF443090

KF443130

KF443128

KF443132

cytb

KF443104

KF443102

KF443106

CO1

GenBank accession no.

646 B.R. Ferrari and G.A.R. Melo

Deceiving colors: recognition of color morphs as separate species in orchid bees

647

Figure 4. Map showing the known distribution of Euglossa townsendi (including the taxa described as E. anodorhynchi and E. aratingae) based on specimens deposited in the entomological collection of the Universidade Federal do Paraná.

servers (Miller et al. 2010) with MrBayes 3.2.1 (Ronquist et al. 2012) from a six-partition data set, divided by gene and codons, with the selection of a mixed evolutionary model instead of an a priori model (Ronquist and Huelsenbeck 2003). A Markov Chain Monte Carlo (MCMC) was randomly run with an initial number of 6×106 generations. After the analysis, 25 % of the trees were discarded, producing a consensus tree with posterior probabilities as a branch support. The maximum likelihood and bootstrap analyses were also conducted through CIPRES servers, with the option of automatic bootstrap stop, from the same sequence alignment used in the Bayesian analysis, by RAxML BlackBox (Stamatakis et al. 2008).

3. RESULTS From the initial 42 specimens, 30 were successfully amplified for the CO1 fragment

and 37 for the cytb fragment, providing 22 samples with both genes amplified and 8 with only one gene amplified (seven for cytb and one for CO1). Three samples (BFt10, BFt12, and BFs03) could not be amplified for any of the genes, possibly due to poor specimen preservation. Three amplifications (BFi11/ CO1, BFi11/cytb, and BFt03/CO1) were contaminated by fungus, as checked using a basic local alignment search tool (BLAST; Altschul et al. 1997). Additionally, ten other samples were compromised by the occurrence of pseudogenes, even after re-amplification and sequencing. Pseudogenes were detected by the presence of stop codons along the sequence, using the Geneious R6 translation tool. Therefore, from the 42 initial samples, 30 were successful sequenced and used in the analysis (Table I).

648

B.R. Ferrari and G.A.R. Melo

Table II. List of additional species of Apini included in this study, with respective published reference and GenBank accession numbers Taxa

Reference

CO1

cytb

Apis mellifera Linnaeus Apis cerana Fabricius Melipona bicolor Lepeletier Bombus hypocrita Perez Bombus ignitus Smith Euglossa amazonica Dressler Euglossa heterosticta Moure

Crozier and Crozier (1993) Tan et al. (2011) Silvestre et al. (2008) Hong et al. (2008) Cha et al. (2007) Ramirez et al. (2010) Ramirez et al. (2010), Darveau et al. (2005) Ramirez et al. (2010) Ramirez et al. (2010), Darveau et al. (2005) Ramirez et al. (2010) Ramirez et al. (2010) Darveau et al. (2005) Darveau et al. (2005) Ramirez et al. (2010), Koulianos et al. (1999) Ramirez et al. (2010), Darveau et al. (2005) Ramirez et al. (2010), Cameron and Mardulyn (2001) Ramirez et al. (2010) Ramirez et al. (2010), Darveau et al. (2005)

NP008083 YP0037355170 NP775708 YP003519421 YP001974606 EU421466 EU421500

NP008093 YP003735180 NP775718 YP002519431 YP001974616 – AY916102

EU421486 EU421573

– AY916113

EU421527 EU421488 – – EU421537

– – AY916115 AY916114 AF002730

EU421478

EU421478

EU421512

AF181614

EU421542 EU421518

– AY916093

Euglossa truncata Rebêlo & Moure Euglossa bursigera Moure Euglossa Euglossa Euglossa Euglossa Euglossa

laevicincta Dressler annectans Dressler chalybeata Friese ignita Smith imperialis Cockerell

Exaerete frontalis (Guérin) Eulaema meriana (Olivier) Aglae caerulea Lepeletier & Serville Eufriesea ornata (Mocsáry)

Phylogenetic reconstructions based on maximum likelihood and Bayesian inference exhibited similar topologies. The consensus tree resulting from the Bayesian analysis, with its posterior probability support over the branches and with the maximum likelihood bootstrap values beneath them, is shown in Figure 5. For illustration purposes, due to the long branches involved, the terminals belonging to other Apini subtribes are not shown. As can be seen in Figure 5, for the two cases involving E. iopoecila and E. stellfeldi, the consensus tree showed very little or no internal topology resolution, with all terminal taxa collapsed together. In both cases, there is no support for separation between green and blue forms. A single specimen of E. stellfeldi, BFs04, showed a disproportionally longer branch, compared to the other terminals. This

blue specimen from Paraná had a maximum uncorrected distance of 1.1 % with the remaining terminals of E. stellfeldi. In the E. townsendi case, some internal resolution has been recovered, with the blue morph samples (BFa01 and BFa02, corresponding to the taxon described as E. anodorhynchi) and the Mexican specimens (BFt07 and BFt08) being more closely related to each other, respectively. This seems simply a result of these samples being more similar to each other, when compared to the remaining terminals, due to their close geographic proximity (Figure 3). In any case, here also, no resolution indicative of separate lineages was evidenced in the phylogenetic analysis. Within the studied cases, average pairwise genetic distance was 0.31 % (±0.18) for E. iopoecila, 0.28 % (±0.40) for E. stellfeldi, and

Deceiving colors: recognition of color morphs as separate species in orchid bees

649

Figure 5. Consensus tree resulting from a Bayesian analysis of molecular data from the genes CO1 and cytb for the three case studies in the genus Euglossa. Posterior probability support is indicated over the branches and the maximum likelihood bootstrap values beneath them.

0.94 % (±0.43) for E. townsendi. Pairwise distances never exceeded 1.7 %, with the

highest value observed within E. townsendi. Comparisons between different colored individ-

650

B.R. Ferrari and G.A.R. Melo

uals in E. townsendi revealed smaller genetic distances (e.g., 0.5 % between BFa01 and BFt02) than similarly colored individuals (e.g., 1.7 % between BFt07, BFt08, or BFt09 and BFt01), suggesting that both morphs should belong to a single taxonomic unit. The two blue specimens of E. townsendi differed from the remaining green individuals by an average distance of 0.82 % (±0.18) and the two Mexican specimens from the remaining ones by 1.21 % (±0.33). 4. DISCUSSION Increased popularization of DNA amplification and sequencing techniques has shed new light on species limits. Many studies have attempted to measure the amount of genetic divergence indicative of independent evolutionary lineages traditionally recognized as separate species, sometimes called the “bar code gap” when referring to divergence in the CO1 gene (Meyer and Paulay 2005). Hebert et al. (2003), when studying Lepidoptera, noticed that 98 % of morphologically valid species presented genetic distances greater than 3 %, and the remaining 2 % were from congeneric individuals, with clear morphological distinction, despite their lower genetic distances, suggesting more recent divergence. Among many metazoan groups, Meier et al. (2008) estimated a mean intraspecific variation of 1.8 % for Hymenoptera, in contrast to a minimum distance of 3.8 % (±4.2) between different species. For orchid bees, there is no study comparing the amount of genetic variation within species with that among different species of a given group, in particular, involving taxa whose divergence time spans over a continuous range from more recent to older divergences. Estimates of intraspecific variation were published by Dick et al. (2004), based on a fragment of the CO1 gene, for 14 non-closely related euglossine species, with values ranging from 0.0 to 2.0 %. Only for a single pair of taxa, involving Euglossa mixta and Euglossa cognata, their CO1 data failed to sort the specimens into the two species recognized through mor-

phological features. However, the group to which these two species belong, the analis group, contains cryptic undescribed species (Melo, unpublished data) and, judging from the collecting data, some of the specimens studied by Dick et al. (2004) belong to these undescribed taxa. More recently, Eltz et al. (2011) also showed that sequence data from a CO1 fragment could not resolve the relationships between two closely related forms in Euglossa, treated by them as separate species. In their case, the two species were distinguished based in the chemical composition of the perfumes found in the hind tibia as well as microsatellite genotyping and the number of mandibular teeth of males. The divergence values for the two mitochondrial genes found here among individuals within each of the case studies, all below 2.0 %, are compatible with those expected for a variation within a single species. Indeed, the average values within E. iopoecila and E. stellfeldi, both around 0.3 %, are very small and reveal a reduced differentiation among populations stretching for over 1,500 km along the coastal forests of eastern Brazil. Even the higher values observed within E. townsendi, as, for example, that of 1.2 % between the Mexican specimens and those from eastern Brazil, are relatively small, considering that the samples came from sites about 7,000 km apart. Despite these results, one could argue that the mitochondrial genes are not adequate for sorting recently diverged lineages in orchid bees as found by Eltz et al. (2011) for the CO1 gene. Differently from their study, in which only the CO1 gene was sequenced and the two forms could be distinguished by a structural character, we used an additional mitochondrial gene, and the forms here studied cannot be morphologically differentiated, except for their integumental color. In this respect, it is important also to emphasize that the blue morphs of the three studied species all occur in the same region, being mostly sympatric in the southern portion of the Atlantic forest. As documented by Faria and Melo (2007), this distribution pattern points to strong selective pressure favoring blue color

Deceiving colors: recognition of color morphs as separate species in orchid bees

phenotypes in orchid bees of the genus Euglossa present in this region. Therefore, we conclude that the recognition of separate taxa for the different color morphs is not supported by the molecular evidence. The current evidence strongly suggests that taxonomic delimitation of orchid bee species based only on general integument coloration does not correctly reflect interspecific relations within the genera. Taking into consideration the potential intraspecific color variation, taxonomists should use these characters secondarily. Continued investigations concerning factors that lead to these color variations and the latitudinal pattern of their distributions are necessary. ACKNOWLEDGMENTS We would like to thank Thomas Eltz for providing the specimens of E. townsendi from Mexico, Marcos Barbeitos for his help with the maximum likelihood analysis, and Kevin A. Williams for his suggestions to an earlier version of the manuscript. GARM thanks Vitor Becker for granting access and for helping during the collection trip to RPPN Serra Bonita and Marco A. Costa for his help during collection trips to Bahia. ICMBio is thanked for the collecting permits and CNPq for financial support (grant 304053/2012-0).

Couleurs trompeuses: la reconnaissance de morphes différant par la couleur comme des espèces séparées chez les abeilles à orchidées n’a pas de fondement moléculaire Forêt atlantique / Brésil / Euglossini / ADN mitochondrial / taxonomie moléculaire

Trügerische Farben: die Anerkennung von Prachtbienen-Farbmorphen als getrennte Arten wird durch molekulare Daten nicht unterstützt Atlantischer Wald / Euglossini / mitochondriale DNA / molekulare Taxonomie

REFERENCES Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman, D.J. (1997) Gapped

651

BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402 Cameron, S.A., Mardulyn, P. (2001) Multiple molecular data sets suggest independent origins of highly eusocial behavior in bees (Hymenoptera: Apinae). Syst. Biol. 50(2), 194–214 Cha, S.Y., Yoon, H.J., Lee, E.M., Yoon, M.H., Hwang, J.S., Jin, B.R., Han, Y.S., Kim, I. (2007) The complete nucleotide sequence and gene organization of the mitochondrial genome of the bumblebee, Bombus ignitus (Hymenoptera: Apidae). Gene 392(1–2), 206–220 Crozier, R.H., Crozier, Y.C. (1993) The mitochondrial genome of the honeybee Apis mellifera: complete sequence and genome organization. Genetics 133, 97–117 Darveau, C.A., Hochachka, P.W., Welch Jr., K.C., Roubik, D.W., Suarez, R.K. (2005) Allometric scaling of flight energetics in Panamanian orchid bees: a comparative phylogenetic approach. J. Exp. Biol. 208(PT 18), 3581–3591 Dick, C.W., Roubik, D.W., Gruber, K.F., Bermingham, E. (2004) Long-distance gene flow and crossAndean dispersal of lowland rainforest bees (Apidae: Euglossini) revealed by comparative mitochondrial DNA phylogeography. Mol. Ecol. 13, 3775– 3785 Dodson, C.H., Dressler, R.L., Hills, H.G., Adams, R.M., Williams, N.H. (1969) Biologically active compounds in orchid fragrances. Science 164, 1243–1249 Dressler, R.L. (1982a) New species of Euglossa II. (Hymenoptera: Apidae). Rev. Biol. Trop. 30, 121–129 Dressler, R.L. (1982b) New species of Euglossa. IV. The cordata and purpurea species groups (Hymenoptera: Apidae). Rev. Biol. Trop. 30(2), 141–150 Eltz, T., Fritzsch, F., Pech, J.R., Zimmermann, Y., Ramírez, S.R., Quezada-Euan, J.J.G., Bembé, B. (2011) Characterization of the orchid bee Euglossa viridissima (Apidae: Euglossini) and a novel cryptic sibling species, by morphological, chemical, and genetic characters. Zool. J. Linn. Soc. 163, 1064 –1076 Faria, L.R.R., Melo, G.A.R. (2007) Species of Euglossa (Glossura) in the Brazilian Atlantic forest, with taxonomic notes on Euglossa stellfeldi Moure (Hymenoptera: Apidae: Euglossina). Rev. Bras. Entomol. 51, 275–284 Folmer, O., Black, M., Hoeh, W., Lutz, R., Vrijenhoek, R. (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol. Mar. Biol. Biotech. 3, 294–297 Hebert, P.D.N., Cywinska, A., Ball, S.L., deWaard, J.R. (2003) Biological identifications through DNA barcodes. Proc. R. Soc. Lond. B. 270, 313–321 Hong, M.Y., Cha, S.Y., Kim, D.Y., Yoon, H.J., Kim, S.R., Hwang, J.S., Kim, K.-G., Han, Y.S., Kim, I. (2008) Presence of several tRNA-like sequences in the mitochondrial genome of the bumblebee,

652

B.R. Ferrari and G.A.R. Melo

Bombus hypocrita sapporoensis (Hymenoptera: Apidae). Genes Genom. 30(4), 307–318

(Apidae, Euglossinae). Rev. Bras. Zool. 12, 445– 466

Katoh, K., Misawa, K., Kuma, K., Miyata, T. (2002) MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30, 3059–3066

Ronquist, F., Huelsenbeck, J.P. (2003) MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572–1574

Koulianos, S., Schmid-Hempel, R., Roubik, D.W., SchmidHempel, P. (1999) Phylogenetic relationships within the corbiculate Apinae (Hymenoptera) and the evolution of eusociality. J. Evol. Biol. 12(2), 380–384 Meier, R., Zhang, G., Ali, F. (2008) The use of mean instead of smallest interspecific distances exaggerates the size of the “barcoding gap” and leads to misidentification. Syst. Biol. 57(5), 809–813 Meyer, C.P., Paulay, G. (2005) DNA barcoding: Error rates based on comprehensive sampling. PLoS Biol. 3, 2229–2238 Miller, M.A., Pfeiffer, W., and Schwartz, T. (2010) Creating the CIPRES Science Gateway for inference of large phylogenetic trees, in: Proceedings of the Gateway Computing Environments Workshop (GCE), New Orleans, pp. 1–8. Nemésio, A. (2006) Euglossa anodorhynchi sp. n. (Hymenoptera: Apidae), a new orchid bee from southern Brazil. Syst. Entomol. 35(2), 206–209 Nemésio, A. (2009) Orchid bees (Hymenoptera: Apidae) of the Brazilian Atlantic Forest. Zootaxa 2041, 1–242 Ramirez, S.R., Roubik, D.W., Skov, C., Pierce, N.E. (2010) Phylogeny, diversification patterns and historical biogeography of euglossine orchid bees (Hymenoptera: Apidae). Biol. J. Linn. Soc. Lond. 100(3), 552–572 Rebêlo, J.M.M., Moure, J.S. (1996) As espécies de Euglossa Latreille do nordeste de São Paulo

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M.A., Huelsenbeck, J.P. (2012) MrBayes 3.2: Efficient Bayesian Phylogenetic Inference and Model Choice Across a Large Model Space. Syst. Biol. 61(3), 539–542 Silvestre, D., Dowton, M., Arias, M.C. (2008) The mitochondrial genome of the stingless bee Melipona bicolor (Hymenoptera, Apidae, Meliponini): Sequence, gene organization and a unique tRNA translocation event conserved across the tribe Meliponini. Genet. Mol. Biol. 31, 451–460 Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H., Flook, P. (1994) Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved PCR primers. Ann. Entomol. Soc. Am. 87, 651–701 Stamatakis, A., Hoover, P., Rougemont, J. (2008) A Rapid Bootstrap Algorithm for the RAxML WebServers. Syst. Biol. 75(5), 758–771 Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S. (2011) MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739 Tan, H.W., Liu, G.H., Dong, X., Lin, R.Q., Song, H.Q., Huang, S.Y., Yuan, Z.G., Zhao, G.H., Zhu, X.Q. (2011) The complete mitochondrial genome of the Asiatic cavity-nesting honeybee Apis cerana (Hymenoptera: Apidae). PLoS ONE 6(8), E23008